seismic splitbam

Purpose

seismic splitbam splits a multi-reference BAM file into one BAM file per reference sequence. Use it when you aligned reads to a multi-reference FASTA using an external aligner and need separate per-reference BAMs before running seismic idmut. (seismic align produces per-reference BAMs automatically; this step is only needed for externally produced BAMs.)

Requires Bowtie2 (≥ 2.5) and SAMtools as external dependencies.

Inputs

Reference FASTA file

First positional argument.

BAM / SAM / CRAM files

One or more alignment files containing reads mapped to multiple references. See Specify Input Files.

Outputs

All outputs go into {out}/{sample}/align/.

{ref}.bam

One BAM file per reference, sorted and indexed.

split-report.json

Summary of settings and results.

Quick example

Split a multi-reference BAM:

seismic splitbam ref.fa multi.bam

Options

Alignment filters
--min-mapq N

Discard reads with mapping quality below N (default 25).

--min-reads N (-N)

Discard BAM files with fewer than N reads (default 1000).

--bt2-orient {fr|rf|ff}

Expected paired-end mate orientation (default fr).

--bt2-X N

Maximum paired-end alignment length (default 600).

Strand separation
--sep-strands/--mix-strands

Split each BAM into forward- and reverse-strand reads (default off).

--rev-label LABEL

Suffix for reverse-strand reference names (default -rev).

Branches
--branch NAME (-b)

Write outputs to {out}/{sample}/align_{NAME}/. See Branches.

Performance

--num-cpus N — multiprocessing; see Parallelize Tasks. --force — overwrite existing outputs.

The auto-generated Command Line Reference lists every option with its current default.

See also